It has become apparent during the previous twenty years that many neurological disease traits do not result from coding region mutations within genes, but instead manifest because of alterations of the genome. Genomic disorders are a class of conditions that result from genomic rearrangements rather than base pair changes of DNA sequence. In general, two major types of rearrangements are observed: recurrent and nonrecurrent genomic changes. Recurrent rearrangements have a common size in different patients; in which the breakpoints occur at 'fixed' genomic positions, or breakpoint cluster regions. The breakpoints cluster in paralogous segments of the human genome (also referred to as low-copy repeats, LCRs, or segmental duplications, SDs) that facilitate a non-allelic homologous recombination (NAHR) by both stimulating and mediating the rearrangement. Nonrecurrent rearrangements can be of different sizes in different patients, but usually share a smallest region of overlap (SRO) in which the critical genomic contents and/or gene(s) reside. Genomic disorders produced by non-recurrent rearrangements provide a unique challenge for studies of genotype/phenotype correlations. The variable size and genomic content as well as the frequent co- occurrence of complex alterations (e.g. triplications and inversions) that can occur with nonrecurrent rearrangements add further complexity to interpreting the genome and gene variation in the context of each patient's clinical manifestations. We hypothesize that nonrecurrent rearrangements may occur by mechanisms that are distinct from homologous recombination mechanisms; our preliminary studies and recent publications from the first year of this stimulus grant strongly support this hypothesis. Furthermore, we suggested some nonrecurrent rearrangements may result because of specific genome architectural features causing susceptibility to such rearrangements. We plan to investigate these hypotheses in an attempt to learn the rules for mechanisms leading to nonrecurrent rearrangements. We will investigate these hypotheses experimentally by; 1) mapping breakpoints of duplication rearrangements, triplication rearrangements, and complex rearrangements. 2) performing bioinformatic analyses of the genomic region undergoing rearrangement, and 3) determining the products of recombination through direct DNA sequences of the recombinant junction; i.e. breakpoint sequencing. 4) studying marker genotypes by whole-genome arrays in trios that consist of patients with disease associated de novo complex rearrangements and their unaffected parents to surmise strand exchanges or potential template switches by the segregation of marker haplotypes. Finally, we will attempt to elucidate genes that may be important to these rearrangement processes by whole genome sequencing of personal genomes in subjects, or parents of these subjects, with multiple de novo CNV events. In this manner we will identify the substrates for recombination, gain insights into genome architecture in regions involved, and potentially infer mechanisms for the rearrangements.